2014 Kellogg School of Management/Aspen Institute Business and Society Leadership Summit
Filling the Governance Gap: Aligning Enterprise and Advocacy #CorpGovGap
Protect Your Global Supply Chain from Costly Disruptions
By Sunil Chopra and ManMohan S. Sodhi
Supply chains, once confined to a single country or continent, have stretched around the world as companies seek to source quality goods at the lowest cost. While managers have become adept at addressing recurrent risks—frequent, low-impact incidents such as demand fluctuations or supply delays that affect efficiency—they also admit to doing very little about preventing or mitigating the impact of disruptive risks such as labor strikes, political unrest, regulatory shifts, and natural disasters.
In the fluid environment found in many countries, the only certainty is that disruptive events will happen; only the timing and magnitude are unknown. An incident that occurred in July 2012 at India's top auto manufacturer, Maruti-Suzuki India Limited (MSIL), is a sobering example: while protesting the suspension of a worker who had attacked his supervisor, a group of MSIL workers reportedly ran amok, killing a senior HR executive, seriously injuring 50 others, and burning down the administrative wing. In addition to the loss of human life, such unforeseen and high-impact events threaten to shut down entire sections of the supply chain and can have a domino effect on other sections across the globe, affecting stock price, market share, capacity, reputation, and more.
But solutions to reduce risk mean little unless they are contextualized within performance improvement or at least the absence of performance deterioration. How should managers prepare for disruptive risks while maintaining gains from improved supply chain efficiency?
Reducing Risk While Improving Performance
To better understand how companies can improve performance while reducing risk, it is helpful to think of supply chain performance and risk as a matrix (Figure 1). The quadrant with higher risk and lower performance may well be the status quo for some supply chains but is certainly not the goal for any. As such, it is the other three quadrants that merit discussion:
Quadrant I: Single-minded increase in performance
Since the mid 1990s, managers have become much better at managing increasingly profitable global supply chains. In the process of building a lean supply chain with centralized inventories, concentrated suppliers, and common parts, however, management has achieved cost efficiency by becoming more vulnerable to disruptive risks. As the experience of global companies in the 21st century has already demonstrated, managerial efforts need to shift from this quadrant to the other two.
Quadrant II: Win–win through "containing" supply chain scope
Complexity is a natural element of juggling the huge number of commodities flowing through a supply chain. "Containing" the supply chain scope can lead to higher performance and lower risk by limiting disruptions to just one section of the supply chain. Two strategies can reduce fragility while improving financial performance:
- Segment your supply chains
Geographic concentration of suppliers keeps transportation costs low and ensures high availability of supplies and labor; however, it also means that all companies in the region can be affected by the same labor problems or regulatory shifts. Indeed, when MSIL fired 500 workers suspected of participating in the violence at the Mansear plant, unions at other companies in the region went on strike in protest.
Large companies can use segmentation to improve profits and reduce supply chain fragility. For high-volume commodity items with low demand uncertainty, the supply chain should have specialized and decentralized capacity. In contrast, for low-volume products with high demand uncertainty, supply chains should be flexible with capacity that is centralized to pool demand. (Figure 2)
As one example, Amazon has increased its number of U.S. distribution centers to be closer to the consumer. Inventories are segmented: the fast-moving items are carried in all distribution centers to reduce transportation cost and improve responsiveness, and the slow-moving items are kept relatively centralized. This segmentation helps reduce costs and fragility.
- Regionalize your supply chains
The concept of containment also applies to regionalizing supply chains. While "globalizing" the supply chain by moving production to lower-cost countries has been popular for the past couple of decades, doing so has increased fragility because the impact of disruptions is felt across the entire supply chain.
Regionalization provides an opportunity to reduce both the fragility of sprawling global supply chains and transportation costs; it achieves the former by containing the impact of a disruptive event to a single region. For example, MSIL’s existing supply chain has two plants and most suppliers in the same area, thus concentrating labor-strife risk. The company has committed to bringing another plant in a different part of the country online in 2015. This additional plant will decrease the impact of a possible labor-related disruption while simultaneously reducing the costs of access to sea for export and transporting cars to other parts of India.
Quadrant III: Trade off a little performance for a large reduction in risk
The reason that executives are reluctant to deal with risk is the (only partly true) perception that steps to mitigate risk will cause a dramatic decline in performance. However, we propose two strategies that can minimize the loss of performance while substantially lowering supply chain risk.
- Overestimate the likelihood of a disruption
Humans have the tendency to underestimate the probability of rare events. Given the upfront cost of protecting a supply chain against disruptive risks, it’s common for managers to downplay or even completely ignore the likelihood that such an incident will occur. Our research indicates that underestimating the probability or duration of a disruption is far more expensive in the long run (in terms of the average cost of many disruptions occurring over an extended period of time) than overestimating the likelihood of a disruptive event.
Our research found that an error in estimating the probability of a disruption by as much as 50 percent results in a less than 2 percent increase in total costs stemming from disruptive risk incidents over the long run. Therefore, companies would be far better served by investing in efforts to shore up a supply chain (for instance, by building additional distribution centers) rather than doing nothing and hoping for the best.
- Reduce concentration of resources
Many managers use centralized inventory or common parts to lower the cost of mitigating recurrent supply chain risks. However, extensive pooling makes the supply chain more vulnerable to disruptive risk. The billions of dollars that Toyota lost in sales and incremental costs as a result of its recall in February 2010, for example, represent the fragility of a supply chain that used a part common in too many car models and sourced from a single supplier.
Management must recognize that dealing with recurrent risks tugs in the direction of more pooling, but dealing with rare, disruptive risks pulls the balance in the opposite direction. Even when the cost is high or economies of scale are serious enough to warrant a single source, companies should avoid adopting extreme levels of concentration. Having one supplier is risky, but retaining ten suppliers increases cost without reducing fragility.
To find an equilibrium, managers must evaluate the cost of increasing or decreasing inventory, capacity, flexibility, responsiveness, and capability, and of centralizing or decentralizing inventory or capacity. For these reasons, large companies such as Samsung Electronics always seek to have a second supplier.
And When Disaster Strikes…
Sodhi and Tang1 have categorized three stages of response: (1) detecting the disruption, (2) designing a solution or selecting a predesigned solution, and (3) deploying the solution. An effective plan must be tailored to the relevant supply chain quadrant to reduce both the response and recovery times.
To reduce detection time, many companies have implemented IT systems for monitoring the flow of materials (delivery and sales) and information (demand forecasts, production schedule, inventory level, quality) on a regular basis. Such systems typically issue alerts should something out of the ordinary occur. For efficient but far-flung supply chains in Quadrant I or III, detection is much harder than for those with contained scope in Quadrant II.
A company and its partners can lower design times by developing contingent recovery plans for different types of disruptions in advance—for example, though business continuity efforts. Li & Fung has a variety of different contingent supply plans to enable it to switch to suppliers in different countries. Quadrant III supply chains have more options to recover, as there will be backup supply sources for critical parts or commodities. For Quadrant II, another segment or region can serve as backup. But for Quadrant I, there are no readily available options.
Finally, improving communication and coordination within the company as well as among supply chain partners can significantly reduce deployment time. Once a plan is put in place, centralized command becomes less necessary as each party has preassigned responsibilities. Where there is a backup as a result of Quadrant II or Quadrant III efforts, a solution can be readily deployed. But for Quadrant I supply chains, a backup supply will have to be identified and qualified before any orders can be made.
The simultaneous pursuit of increased performance and resilience in the face of disruption are not necessarily in opposition. By containing the impact of disruptions in certain segments and regions, recognizing the true costs of such disruptions, and taking targeted action to infuse the right level of redundancy into the supply chain, companies can be well positioned to weather any storm.
1Sodhi and Tang (2009), Sodhi and Tang (2012): Chapter 5